This proposal presents an approach to the understanding of the catalytic mechanisms of several important enzymes in nucleotide metabolism. The key feature of the approach is the utilization of prudently designed stable analogs of postulated enzyme-generated intermediates. Potent inhibition of the enzyme activity by these analogs is an indication of the presence along the reaction coordinate, of the intermediate after which the analog was modeled.